Assessing the quality of identified models through the asymptotic theory -- when is the result reliable?

DOI10.1016/j.automatica.2004.03.005zbMath1073.93013OpenAlexW2140687796MaRDI QIDQ705145

Simone Garatti, Marco C. Campi, Bittanti, Sergio

Publication date: 26 January 2005

Published in: Automatica (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.automatica.2004.03.005

zbMATH Keywords

System identificationBox-JenkinsAsymptotic theoryARMAXModel quality assessment

Mathematics Subject Classification ID

System identification (93B30) Identification in stochastic control theory (93E12)

Related Items

Identification of multi-input systems: variance analysis and input design issues ⋮ Necessary and sufficient conditions for uniqueness of the minimum in Prediction Error Identification ⋮ Non-asymptotic model quality assessment of transfer functions at multiple frequency points ⋮ Use Hausdorff metric to analyze convergence of parameter estimation in system identification ⋮ A counterexample to the uniqueness of the asymptotic estimate in ARMAX model identification via the correlation approach ⋮ Identification of symmetric noncausal processes ⋮ Parameter identification for nonlinear systems: guaranteed confidence regions through LSCR ⋮ Covariance analysis in SISO linear systems identification ⋮ Iterative robust control: speeding up improvement through iterations ⋮ Asymptotic statistical analysis for model-based control design strategies ⋮ The nonasymptotic confidence set for parameters of a linear control object under an arbitrary external disturbance ⋮ Non-asymptotic quality assessment of generalised FIR models with periodic inputs ⋮ On resampling and uncertainty estimation in linear system identification ⋮ Bayesian system identification via Markov chain Monte Carlo techniques ⋮ The asymptotic model quality assessment for instrumental variable identification revisited ⋮ Non-asymptotic confidence regions for the parameters of EIV systems ⋮ Non-asymptotic confidence regions for model parameters in the presence of unmodelled dynamics ⋮ Guaranteed non-asymptotic confidence regions in system identification ⋮ Perturbed datasets methods for hypothesis testing and structure of corresponding confidence sets

Cites Work

• Unnamed Item
• Unnamed Item
• On covariance function tests used in system identification
• Comparing different approaches to model error modeling in robust identification
• Non-asymptotic confidence ellipsoids for the least-squares estimate
• Identification and control -- Closed-loop issues
• Estimation of model quality
• Special issue on trends in system identification
• Estimating linear time-invariant models of nonlinear time-varying systems
• Asymptotic variance expressions for identified black-box transfer function models
• Quantifying the error in estimated transfer functions with application to model order selection
• Estimating model variance in the case of undermodeling
• Convergence analysis of parametric identification methods
• The role of model validation for assessing the size of the unmodeled dynamics
• Identification of probabilistic system uncertainty regions by explicit evaluation of bias and variance errors
• The fundamental role of general orthonormal bases in system identification
• Asymptotic variance expressions for estimated frequency functions
• Finite sample properties of system identification methods
• Using the bootstrap to estimate the variance in the case of undermodeling